Item description for Evolutionary Algorithms: The Role of Mutation and Recombination (Natural Computing Series) by William M. Spears...
Despite decades of work in evolutionary algorithms, there remains a lot of uncertainty as to when it is beneficial or detrimental to use recombination or mutation. This book provides a characterization of the roles that recombination and mutation play in evolutionary algorithms. It integrates prior theoretical work and introduces new theoretical techniques for studying evolutionary algorithms. An aggregation algorithm for Markov chains is introduced which is useful for studying not only evolutionary algorithms specifically, but also complex systems in general. Practical consequences of the theory are explored and a novel method for comparing search and optimization algorithms is introduced. A focus on discrete rather than real-valued representations allows the book to bridge multiple communities, including evolutionary biologists and population geneticists.
Promise Angels is dedicated to bringing you great books at great prices. Whether you read for entertainment, to learn, or for literacy - you will find what you want at promiseangels.com!
Est. Packaging Dimensions: Length: 9.54" Width: 6.41" Height: 0.69" Weight: 1.05 lbs.
Release Date Sep 20, 2004
ISBN 3540669507 ISBN13 9783540669500
Availability 101 units. Availability accurate as of May 28, 2017 02:50.
Usually ships within one to two business days from La Vergne, TN.
Orders shipping to an address other than a confirmed Credit Card / Paypal Billing address may incur and additional processing delay.
Reviews - What do customers think about Evolutionary Algorithms: The Role of Mutation and Recombination (Natural Computing Series)?
Interesting. Very academic Jun 13, 2001
This book is based on the author's PhD dissertation and it shows (you can download the dissertation from the web). There is page after page of mind numbing step-by-step derivations that do not add too much to the discussion. I would have enjoyed the book more if Spears had shortened some of his derivations. I found the results interesting. Although some of the conclusions seem fairly obvious after reading the book, I think it is important that someone took the time to come up with the mathematical models to formalize things.
The empirical approach is very interesting, and I wish more people would follow and improve on Spears' ideas. Empirical studies of evolutionary algorithms are justly critized for being too limited to a few "standard" functions that do not show much about the capabilities and limitations of the algorithms. Spears took a good step in emulating the machine learning comunity and using test problem generators. With these generators, the experimenters can play around with parameters such as the multimodality or noise in a problem and make systematic empirical studies of the algorithms. Unfortunately, it is difficult to translate from those systematic studies to real life. For example, how much noise or how many peaks are in real-life problems?
Still, I would recommend to go and read this book (or the free dissertation). Skip the equations, though, and get to the point.
BTW, Dr Gordon (the first reviewer) is married to Spears, which may explain some of the excitement in her review...
Essential Reading on Evolutionary Algorithms Dec 24, 2000
This book is an essential resource for anyone studying the theoretical underpinnings of evolutionary algorithms (EAs). The book very carefully analyzes the effects of two fundamental evolutionary operators, recombination and mutation, and their interaction with evolutionary selection. This analysis significantly enhanced my understanding of EAs because of the fundamental role that these operators play. The book begins with the more traditional static analysis approach, but soon it transitions to a very exciting dynamic analysis. Just as neurophysiologists have discovered that when studying the brain it helps to view it as a dynamic process, Spears illustrates how much better we can understand EAs when using dynamic models, such as the popular Markov chain model approach. One of the best parts of the book was the creative use of problem generators for empirically testing the theory and for characterizing the classes of problems for which each EA operator is more effective. This was exciting for two reasons. For one, it encourages EA researchers to break away from myopic use of the same old test suites. Secondly, the problem characterization has tremendous potential value for practical applications of EAs.
Another of my favorite parts of the book was Spears' novel algorithm for compressing Markov chains. I particularly liked the mathematical analysis, which was both elegant and clear. Because Markov chains are widely used, e.g., in operations research, control theory, and artificial intelligence, this compression algorithm has wide-reaching implications for reducing the complexity of modeling a variety of systems.
The intended audience for Spears' book is computer scientists, mathematicians, and biologists, as well as students of evolutionary processes. To make the book accessible to such a diverse audience, the presentation is exceptionally clear and devoid of excessive jargon and obscure mathematics. Only an undergraduate level math background is required. One thing that I found mildly distracting was the repetition between chapters. The reason for the repetition was to make the chapters as self-sufficient as possible. Nevertheless, I read the book as a continuous whole and for anyone who does this I recommend skimming or skipping over the redunant portions. If this is done, the reader can maintain a high level of interest.
In conclusion, because of the valuable insights I gleaned from this book I believe it should be required reading for anyone who wishes to gain a better understanding of evolution as simulated by EAs. Spears' rigorous analyses and lucid explanations make this a delightful book to read.